Because of coding inaccuracies, a total of 380 consecutive operative reports were hand-screened to generate the desired sample size. In particular, the orthopedic surgeon who performed the procedure was often different from the attending physician listed by QSHI, and the procedure performed was often different from the procedure code listed by QSHI. Limits of time and physical space (in the medical records department) precluded random sampling of patients treated by each subspecialist. In the 3-month study period, data for equal numbers of total hip and knee replacements were sought for each of the 15 orthopedic surgeons who performed the majority of primary hip and knee procedures in the Calgary Health Region; however, data for 1 surgeon could not be found. Further, for one of the surgeons (identified by letter E), only 5 operative reports for the predefined period could be found. Therefore, a total of 109 charts were reviewed in detail and were included in the analysis.

Figure 1. Types of bone cement

Figure 1. Types of bone cement used by each of 14 orthopedic surgeons (identified by letters A to N).

Sixty-five (60%) of the 109 patients received bone cement that contained antibiotic. Osteoarthritis was common in both the overall sample and the subset of those who received ABC (Table 1). Aggregate QSHI data were used to calculate the mean and median American Society of Anesthesiologists (ASA) score for all patients who underwent primary total hip or knee arthroplasty at all 3 hospital sites during the study period. The ASA score, a preoperative rating of the patient’s general health status and coexisting conditions, ranges from 1 (healthy) to 5 (not expected to survive longer than 24 h)5; both the mean and median scores for all patients were 2 (M. Brandt, Consultant, Health Outcomes, Calgary Health Region; personal communication by telephone, December 21, 2007).

Table 1. Demographic Characteristics of Patients in an Audit of Prophylactic Use of Antibiotic-Containing Bone Cement for Arthroplastic Surgery


No.(%) of Patients*

Characteristic

All Patients (n = 109)

Patients Who Received

Antibiotic Bone Cement (n = 65)


Age (mean and range)


70


(32-91)


69


(48-85)


Sex (female)


71


(65)


36


(55)


Comorbidity


Osteoarthritis


96


(88)


34


(52)


Rheumatoid arthritis


9


(8)


1


(2)


Diabetes mellitus


17


(16)


6


(9)


Chronic renal insufficiency


4


(4)


2


(3)


Inflammatory bowel disease


1


(1)


0


(0)


*Except where indicated otherwise.

Figure 2. Immne-status risk factors

Figure 2. Immne-status risk factors for 65 patients for whom antibiotic- containing bone cement was used in conjunction with hip or knee arthroplasty. CRI = chronic renal insufficiency, C smoker = current smoker, DM = diabetes mellitus, IBD = inflammatory bowel disease, IMM = immunosuppressive medication, N smoker = nonsmoker, P smoker = past smoker, RA = rheumatoid arthritis.

The most frequently selected ABC was a product containing 1 g of tobramycin per 40 g of bone cement (Table 2). Of the 14 surgeons for whom data were available, 11 (79%) used a commercially available ABC. Half of the surgeons (A, C, D, E, H, I, J) consistently used ABC and chose the product containing tobramycin 1 g/40 g (Figure 1). Most (57 or 88%) of the patients who received ABC had either none (23/65) or one (34/65) of the predefined immune-status factors (Figure 2). However, patients with at least one immune-status factor were significantly more likely to receive ABC than patients with no immune-status factors (Fisher’s exact test, 2-tailed p < 0.001). Finally, of the 4 patients who had chronic renal insufficiency, 2 received bone cement containing either gentamicin or tobramycin.

Table 2. Types of Antibiotic Bone Cement Used for Hip and Joint Arthroplasty

Type of Bone Cement

Antibiotic

Concentration of

No.(%) of


(Manufacturer)


Antibiotic (w/w)


Procedures

Containing antibiotic


Simplex P (Stryker, Rutherford, NJ)


Tobramycin



1

g/40 g


53 (49)


DePuy
CMW


1


(Johnson


&


Johnson,


New Brunswick, NJ)


Gentamicin



1

g/40 g


8 (7)


Refobacin Palacos R (Biomet, Warsaw,
IN)


Gentamicin



0.5

g/40 g


3 (3)


Not
containing antibiotic


Palacos R (Zimmer, Inc, Warsaw, IN)


NA


NA


27 (25)


Simplex P (Stryker, Rutherford, NJ)


NA


NA


13 (12)



Multiple products*


NA


NA


1 (1)


Not
stated


NA


NA


4 (4)


*Forone procedure, the surgeon used 2 products: Simplex P containing tobramycin and Simplex P


without antibiotic.